#1244 Strange bad exit status in %install
Closed: Invalid 4 years ago by praiskup. Opened 4 years ago by iucar.

Not sure why this is failing: https://copr.fedorainfracloud.org/coprs/iucar/cran/build/1213619/. The install command seems to work ok, but the log reports a bad exit status. There are a bunch of builds like this in rawhide. Same versions succeeded in Fedora 30 and 31.


Hello @iucar,
thank you for the report.

I don't think this is a Copr issue, but rather a packaging one. Can it be possible?
Try to reproduce the issue locally, without Copr:

curl https://copr-be.cloud.fedoraproject.org/results/iucar/cran/fedora-rawhide-x86_64/01213619-R-CRAN-amber/R-CRAN-amber.spec > /tmp/R-CRAN-amber.spec
spectool -g -R /tmp/R-CRAN-amber.spec
rpmbuild -bs /tmp/R-CRAN-amber.spec
copr-cli mock-config iucar/cran fedora-rawhide-x86_64 > /tmp/iucar-cran-rawhide.cfg
mock -r /tmp/iucar-cran-rawhide.cfg ~/rpmbuild/SRPMS/R-CRAN-amber-0.1.6-1.fc30.src.rpm

IMHO this is either a packaging issue or mock or some lower-level tool issue. I am not sure which, yet.

Yeap, it fails locally, but here's a successful build some days ago of the same spec in Fedora 31, same mock version: https://copr-be.cloud.fedoraproject.org/results/iucar/cran/fedora-31-x86_64/01154907-R-CRAN-amber/builder-live.log.gz. The only difference I can tell is the Python version: 3.7.5 vs. 3.7.6 for the failing builds. The nspawn_args are different too.

I have no idea how to debug this, but it can't be a packaging issue (i.e., the spec is fine, the package installation is correct, but then the return code is 1 without any apparent error). Any ideas? Where's the proper place to report this?

The file ./amber/amber/inst/extdata/ne_50m_admin_0_countries/ne_50m_admin_0_countries.prj doesn't have a new-line, which breaks the brp-mangle-shebangs script. I'll try to have a look
at this next week, if nobody is faster :-)

Thanks, @praiskup, indeed that's the issue. I found it was already reported one month ago: https://bugzilla.redhat.com/show_bug.cgi?id=1787088, but no fix so far. You can close this unless you want to keep track of it here.

Metadata Update from @praiskup:
- Issue close_status updated to: Invalid
- Issue status updated to: Closed (was: Open)

4 years ago

Yeah, bugzilla is enough. Thanks for the report anyways!

Login to comment on this ticket.

Metadata